[en] To examine crack growth with superimposed creep and fatigue stresses, fatigue crack growth tests, with stopping times, tests with ramp loading with different duration periods and tests with successive periods of pure creep and fatigue crack growth were carried out. The investigations were done on the iron-based alloy 800 at 700degC. The crack growth in the stopping time and ramp experiments were described with the parameters of pure creep crack growth and fatigue crack growth based on a linear damage accumulation theory. This damage accumulation theory was also verified in experiments on large samples and pipe geometries with artificial faults. The second part of the report is concerned with experiments on samples of stressed or thermally stored sample material. These experiments are intended to clear up in what way the ageing of the material changes the fracture mechanics properties. To examine their transferability, experiments on creep and fatigue crack growth are carried out on pipes stressed in operation. In general, no differences in crack propagation behaviour are found for material before and after storage. (orig.)

[en] For examining the crack growth under conditions of superimposed creep and fatigue stress, fatigue crack growth tests with hold times (30 sec. up to 4 hours), tests with non-linear loading for various time periods, and tests combining sequential periods of pure creep or pure fatigue stress have been performed with 25 mm CT specimens of the iron base alloy 800 (X10NiCrAlTi 32 20) at an elevated temperature of 700 C. At this temperature, the material exhibits at about equal stress levels approximately equal contributions of creep crack growth and fatigue crack growth. The crack growth behaviour during hold periods and under non-linear stress is described using the characteristics of pure creep crack growth or fatigue crack growth, in the frame of a linear damage accumulation model. (orig.)

[en] To investigate the crack growth behaviour under combined creep-fatigue loading, CT 25 mm-specimens of X10NiCrAlTi 32 20 (Alloy 800) have been tested in experiments with cyclic loadings and hold times, with static loadings and short stress rekief interrupts, with ramp type loadings and with sequences of separate fatigue and creep crack growth periods. The test temperature of 700deg C was selected because only in this temperature range this alloy provides similar amounts of crack growth under creep and fatigue conditions due to equivalent stress levels. For the estimation of crack growth under combined loading conditions a linear accumulation of increase in crack length was proved using the crack growth laws of pure creep and fatigue crack growth. Hold time and ramp loadings lead to a higher crack growth rate compared with pure creep or pure fatigue crack growth tests. In hold time experiments the crack growth rate is higher than ramp tests of the same period time. The results of hold time tests can be fairly enough predicted by linear damage accumulation rules. (orig.)

[en] In order to assess the applicability of results obtained with 1''CT specimens, on fracture mechanical parameters of fatigue crack growth at 700⁰C and 850⁰C (and also in some tests at room temperature), the highly heat resistant alloys X 10 NiCrAlTi 32 20 (INCOLOY 800 H) and NiCr 22 Co 12 Mo (INCONEL 617) have been chosen for further experiments. The specimen geometries are CT 1'', CT 1/2'', and CCP 1/2'', and thickwalled pipes with circumferential defect at outer surface. The loads applied are of the pulsating tensile stress type (R = 0.05), at a frequency of 5Hz. Results reveal that at all temperatures, the fatigue crack growth is independent of specimen geometries, so that the linear-elastic ΔK_I-concept is suitable for describing the crack growth under cyclic load. (orig.)

[en] Investigations on the fracture mechanics behaviour of X10 NiCrAlTi 32 20 (Alloy 800) have been performed at a temperature of 550⁰C. These investigations were part of the safety philosophy for the steam cycle of a power producing HTGR-plant. Emphasis was laid on the demonstration of the transferability of creep and fatigue crack growth laws. The fatigue crack growth data measured with several standard specimens are in good agreement. So it can be assumed, that a description of the fatigue crack growth by means of the linear-elastic parameter is valid. In addition an experiment was performed with a thick walled tube in four point bending, containing a non through wall circumferential crack. By several approximation methods a conservative estimation of the fatigue crack growth in this tube is possible. The creep crack growth in standard specimens is well described by an energy rate integral. The linear elastic description gives a large scatter of data, hence it seems not to be a valid parameter for the description of the creep crack growth. In thin walled heat exchanger tubes with circumferential cracks in tension, the fatigue crack growth very roughly can be described by means of the parameter. This is valid although the formal suppositions are not fulfilled. The failure stress for these tubes in tensile loading is described conservatively by means of the Battelle formula. (orig./HP)

[en] Experiments on the transferability of fracture mechanics data to thick walled tubes have been performed with the materials X10NiCrAlTi 32 20 (Alloy 800 H) and NiCr 22 Co 12 Mo (Inconel 617). The fatigue crack growth is well described by the linear elastic concept up to the highest testing temperatures (900degC). For the description of the creep crack propagation in tubes the parameter 'energy rate integral' seems to be suitable. (orig./HP)

[en] Experiments on the transferability of fracture mechanics data to thick walled tubes have been performed with the materials X10NiCrAlTi 32 20 (Alloy 800 H) and NiCr 22 Co 12 Mo (Inconel 617). The fatigue crack growth is well described by the linear elastic ΔK_I-concept up to the highest testing temperatures (900⁰C). For the description of the creep crack propagation in tubes the parameter C^* seems to be suitable. (orig.)

[en] FCG tests have been carried out with hold times between 30 seconds and 4 hours. Experiments with hold times of more than 32 minutes lead to results similar of pure CCG-tests. FCG-tests with hold times at load maximum show larger crack propagation rates as pure FCG- and pure CCG-tests. The increase of the crack propagation rate with hold times is an expression of the CCG on the FCG. The crack growth rate per cycle is increased if the duration for the loading ramp is increased from 4 to 8 minutes. A linear damage accumulation rule was used to give a mathematical description for the interaction of CCG and FCG. A linear damage accumulation rule can be used to describe the crack propagation in FCG-experiments with hold times. The calculation is based on the laws for pure FCG (da/dN vs. ΔK_I) and pure CCG (da/dt vs. C^*). If the lower borders of pure CCG- and FCG-experiments are used, the results are conservative. Some experiments with saw tooth shaped loading cycles and different load step rates were performed. The attempt to apply the linear damage accumulation rule to these experiments by considering the CCG during the load increase was negative. (author)

[en] Investigations on the fracture mechanics behaviour of X10 NiCrAlTi 32 20 (Alloy 800) have been performed at temperatures of 550⁰C and 700⁰C. These investigations were part of the safety philosophy for the steam cycle of a power producing HTGR-plant. An emphasis was laid on the demonstration of the transferability of creep and fatigue crack growth laws. The fatigue crack growth data measured with several standard specimens are in good agreement. So it can be assumed, that a description of the fatigue crack growth by means of the linear-elastic Δ K_I-parameter is valid. In addition an experiment was performed with a thick walled tube in four point bending, containing a non through wall circumferential crack. By several approximation methods a conservative estimation of the fatigue crack growth in this tube is possible. The creep crack growth in standard specimens is well described by the energy rate integral C^*. The linear-elastic description by means of K_I gives a large scatter of data, hence K_I seems not to be a valid parameter for the description of the creep crack growth. In thin walled heat exchanger tubes with circumferential cracks in tension, the fatigue crack growth very roughly can be described by means of the Δ K_I-parameter. This is valid although the formal suppositions are not fulfilled. The failure stress for these tubes in tensile loading is described conservatively by means of the Battelle formula. (orig.)

[en] The examples listed show that the DC potential method constitutes a method with which crack growth measurements can be effected with sufficient accuracy even on complex geometrical structures. The method is particularly recommendable when the monitoring of many potential measuring points has to take place simultaneously. The results of measurements with various specimen configurations (steam reformer tubes of methane-reforming plant) show that the concept is suitable for describing the growth of fatigue cracks in X10NrCrAlTi 32 20 at temperatures of up to 850⁰C. Creep crack growth in standard samples appears to be describable with the help of an energy rate integral. However, to make a conclusive statement, more experiments will have to be performed. (orig./DG)